A growing number of older adults with dementia are ?unbefriended?: they have impaired capacity and no family or friends to make decisions on their behalf. Because such persons must be represented by a stranger ? most often a guardian, selected by the court ? they may receive care that is discordant with their preferences. Work by the candidate, for example, suggests that individuals with dementia who are under guardianship are much more likely to receive aggressive end-of-life treatment than those with family members available to make decisions. Given the substantial difficulties involved in making decisions for a person with dementia whose values and priorities are unknown, the current application seeks to lay the groundwork for an innovative ?upstream? approach among persons who still have capacity but do not have a potential surrogate, so that they are at risk for becoming unbefriended. The candidate envisions an intervention whereby such persons would be identified ahead of time and health professionals would elucidate their values and priorities. The proposed work will address the key knowledge gaps that stand in the way of such an intervention. Because little is known about the population that would be targeted, Aim 1 will use a unique national dataset to describe the prevalence and risk factors associated with older adults who are unable to name a surrogate. Aims 2 and 3 take up the broader challenge of generating an advance care planning model tailored to the unique challenges of dementia. Aim 2 will involve using qualitative methods to ascertain the core information that shapes treatment decisions when an ideal surrogate is exercising substituted judgment. Aim 3 involves the development and validation of a tool, capturing this information, that can be used in clinical practice. The candidate, Dr. Cohen, is a geriatrician at the Yale School of Medicine with a track record of early success, including several high-impact original reports and a GEMSSTAR award from the NIA. He has engaged an exceptional mentorship team. His primary mentor, Dr. Terri Fried, is an internationally-recognized authority on decision-making for older adults with serious illness. He has recruited three co-mentors and an advisor whose diverse academic backgrounds will contribute a remarkable richness of perspectives and expertise both to the proposed research and to the career development plan. Dr. Cohen has outlined a rigorous program of training that draws upon resources from across Yale University as well as national training opportunities in mixed methods research and leadership development. The extraordinary resources available from the Yale Section of Geriatrics and Program on Aging provide an ideal environment for the execution of the proposed research and for Dr. Cohen?s emergence as an independent investigator at the forefront of geriatrics, medical decision- making, and ethics.